Abstract P390: Cardiometabolic Disease Prevalence Are Inconsistent by Racial Group Across Three National United States Datasets From 2015 to 2021

Abstract

Introduction: The United States uses the National Health Interview Survey (NHIS), Behavioral Risk Factor Surveillance System (BRFSS), and National Health and Nutrition Examination Survey (NHANES) to inform disease trends and preventative research. However, the consistency in prevalence of cardiometabolic disorders in these datasets by race is understudied. We examined differences in the crude and age/sex adjusted prevalence of self-reported doctor diagnosis of myocardial infarction (MI), stroke (CVA), coronary heart disease (CHD), any cardiovascular disease (CVD), and diabetes (DM) by racial group across the datasets. Methods: We harmonized available NHIS (N = 179,593), BRFSS (N = 2,592,050), and NHANES (N = 12,506) datasets of adults aged ≥ 30 from 2015-2021. We first calculated crude prevalence by race and survey. We then used multivariable logistic regression to calculate age and sex adjusted prevalence by race and survey, accounting for survey weights. Results: The majority of participants were female (52.3%) and 53.6 years old (SD=14.4). Crude prevalence was most similar within a given race group for CVA (Table). Prevalence across surveys varied up to 4.6% by race and was most variable on average for DM. For each race group, prevalence estimates were generally higher in BRFSS than NHIS. Prevalence estimates varied most in people of Other/Multiple race. Racial differences in DM were significant in all datasets (p<0.001). Results were similar after adjustment for age and sex. Conclusion: While prevalence estimates for our national datasets are consistent for the NH-White population, estimates are less consistent for other racial groups. Additional work is needed to ensure consistency of cardiometabolic surveillance both nationally and among minoritized groups.